MIMO radar supporting object reflected signal overlap detection

Abstract

A system and method for processing a received radar signal is presented. A distance-Doppler map is determined that includes values associated with the plurality of distance partitions and the plurality of Doppler partitions. A sub-array is determined using the distance-Doppler map. A plurality of frequency spectrums is calculated using the first sub-array. Each spectrum of the plurality of spectrums is associated with a transmit channel of a plurality of transmit channels. An attribute of each of the plurality of spectra is determined. A first spectrum of the plurality of spectrums is determined that includes local peaks not in other spectrums of the plurality of spectrums. A value associated with the transmit channel associated with the first spectrum in the range-Doppler map is modified to determine a corrected range-Doppler map. An estimated direction of arrival of the first object is determined using the corrected distance-Doppler map.

Description

Technical Field

[0001] The present disclosure generally relates to radar systems and associated methods of operation. In one aspect, the present disclosure relates to a Doppler Division Multiplexing (DDM) Multiple-Input Multiple-Output (MIMO) radar system, such as an automotive radar system, configured to mitigate Doppler spectral overlap in signals processed by the radar system to automate certain vehicle functions (e.g., parking assistance and / or potential collision safety alerts). Background Art

[0002] Radar systems, such as automotive radar systems, transmit electromagnetic signals and receive the backscatter of the transmitted signals. The time delay and / or change in time delay between the transmitted signal and the received signal can be determined and used to calculate the distance and / or speed of objects, such as cars, trucks, motorcycles, traffic signs, road fixtures, etc., causing the reflection, respectively. For example, in automotive applications, an automotive radar system can be used to determine the distance and / or speed of oncoming vehicles and other obstacles.

[0003] Automotive radar systems enable the implementation of Advanced Driver Assistance System (ADAS) functions, and are likely to enable safer driving and ultimately a fully autonomous driving platform. Such systems use radar systems as the primary sensors for ADAS operation.

[0004] In a radar system, when determining the Direction of Arrival (DOA) of an object represented in a range-Doppler dataset, signal processing requires analysis of those datasets. If the radar system is configured as a Doppler Division Multiplexing (DDM) Multiple-Input Multiple-Output (MIMO) radar system, there is a risk of Doppler spectral signal overlap within the dataset, which can reduce the accuracy of any such DOA estimate, especially when higher speed or nearby objects are located near the vehicle radar system. Summary of the Invention

[0005] According to a first aspect of the present invention, there is provided a radar system, comprising:

[0006] a plurality of transmitter modules configured to transmit a plurality of transmitted radar signals according to a Doppler domain multiplexing (DDM) scheme, wherein each transmitted radar signal is associated with a transmission channel among a plurality of transmission channels;

[0007] a plurality of receiver modules configured to receive reflections of the plurality of transmitted signals reflected by at least one object and generate signals based on the received reflections; and

[0008] a signal processor configured to:

[0009] Use the signals received by the plurality of receiver modules to determine a range-Doppler map, the range-Doppler map including values associated with a plurality of range bins and a plurality of Doppler bins.

[0010] Determine a first subarray including values associated with a first range bin of the range-Doppler map, wherein the first range bin is associated with a first object.

[0011] Use the first subarray to calculate a plurality of spectra, wherein each of the plurality of spectra is associated with a transmit channel among the plurality of transmit channels.

[0012] Calculate an attribute of each of the plurality of spectra.

[0013] Use the attribute of each spectrum to determine a first spectrum among the plurality of spectra that includes a local peak not in other spectra among the plurality of spectra.

[0014] Modify the value in the range-Doppler map associated with the transmit channel associated with the first spectrum to determine a corrected range-Doppler map; and

[0015] Use the corrected range-Doppler map to determine an estimated direction of arrival of the first object.

[0016] In one or more embodiments, to calculate the attribute of each of the plurality of spectra, the signal processor is configured to determine the number of local peaks within a threshold of the maximum value in each spectrum.

[0017] In one or more embodiments, to use the attribute of each spectrum to determine a first spectrum among the plurality of spectra that includes a local peak not in other spectra among the plurality of spectra, the signal processor is configured to determine that the number of peaks in the first spectrum is different from the number of peaks in each other spectrum among the plurality of spectra.

[0018] In one or more embodiments, the plurality of receiver modules are arranged in a linear receiving subarray, and the first subarray is a uniform receiving subarray.

[0019] In one or more embodiments, to modify the value in the range-Doppler map associated with the transmit channel associated with the first spectrum to determine a corrected range-Doppler map, the signal processor is configured to set the value in the range-Doppler map associated with the transmit channel associated with the first spectrum to a zero value.

[0020] In one or more embodiments, to calculate the plurality of spectra, the signal processor is configured to calculate the fast Fourier transform of the values in the first subarray.

[0021] In one or more embodiments, to compute an attribute of each of the plurality of spectra, the signal processor is configured to compute a cross-correlation value of spectral pairs in the plurality of spectra.

[0022] According to a second aspect of the present invention, there is provided a radar system comprising:

[0023] a transmitter system configured to transmit transmitted signals, wherein each transmitted signal is associated with a transmit channel in a plurality of transmit channels;

[0024] a receiver system configured to receive reflected signals; and

[0025] a signal processor configured to:

[0026] determine a range-Doppler map using the reflected signals,

[0027] determine a first sub-array comprising values associated with a first range partition of the range-Doppler map;

[0028] compute a plurality of spectra using the first sub-array, wherein each of the plurality of spectra is associated with a transmit channel in a plurality of transmit channels;

[0029] determine that the first spectrum includes local peaks not in other spectra of the plurality of spectra by detecting local peaks in the first spectrum of the plurality of spectra; and

[0030] modify values in the range-Doppler map associated with the transmit channel associated with the first spectrum to determine a corrected range-Doppler map.

[0031] In one or more embodiments, to determine that the first spectrum includes local peaks not in other spectra of the plurality of spectra, the signal processor is configured to determine that the number of peaks in the first spectrum is different from the number of peaks in each other spectrum of the plurality of spectra.

[0032] In one or more embodiments, the receiver system includes a plurality of receiver modules arranged in a linear receive sub-array, and the first sub-array is a uniform receive sub-array.

[0033] In one or more embodiments, to compute the plurality of spectra, the signal processor is configured to compute a fast Fourier transform of values in the first sub-array.

[0034] In one or more embodiments, the signal processor is configured to compute a cross-correlation value of spectral pairs in the plurality of spectra.

[0035] According to a third aspect of the present invention, there is provided a method, comprising:

[0036] using signals received by a plurality of receiver modules to determine a range-Doppler map, the range-Doppler map including values associated with a plurality of range bins and a plurality of Doppler bins;

[0037] determining a first sub-array including values associated with a first range bin of the range-Doppler map, wherein the first range bin is associated with a first object;

[0038] using the first sub-array to calculate a plurality of spectra, wherein each spectrum of the plurality of spectra is associated with a transmission channel among the plurality of transmission channels;

[0039] calculating an attribute of each spectrum of the plurality of spectra;

[0040] using the attribute of each spectrum to determine a first spectrum among the plurality of spectra that includes a local peak not in other spectra among the plurality of spectra;

[0041] modifying values in the range-Doppler map associated with the transmission channel associated with the first spectrum to determine a corrected range-Doppler map; and

[0042] using the corrected range-Doppler map to determine an estimated direction of arrival of the first object.

[0043] In one or more embodiments, the method further comprises determining, for calculating the attribute of each spectrum of the plurality of spectra, the number of local peaks within a threshold of the maximum value of each spectrum.

[0044] In one or more embodiments, the method further comprises determining, for using the attribute of each spectrum to determine a first spectrum among the plurality of spectra that includes a local peak not in other spectra among the plurality of spectra, that the number of peaks in the first spectrum is different from the number of peaks in each other spectrum among the plurality of spectra.

[0045] In one or more embodiments, the plurality of receiver modules are arranged as a linear receiving sub-array, and the first sub-array is a uniform receiving sub-array.

[0046] In one or more embodiments, the method further comprises setting, for modifying values in the range-Doppler map associated with the transmission channel associated with the first spectrum to determine a corrected range-Doppler map, the values in the range-Doppler map associated with the transmission channel associated with the first spectrum to a zero value.

[0047] In one or more embodiments, the method further includes calculating a fast Fourier transform of the values in the first subarray in order to calculate the plurality of spectra.

[0048] In one or more embodiments, the method further includes calculating a cross-correlation value of a pair of spectra in the plurality of spectra in order to calculate an attribute of each of the plurality of spectra.

[0049] In one or more embodiments, the method further includes sending an output based on the estimated direction of arrival to an autonomous driving assistance system of a vehicle.

[0050] These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] A more complete understanding of the subject matter can be obtained by reference to the detailed description and claims considered in conjunction with the following drawings, in which like reference numerals refer to like elements throughout the figures.

[0052] Figure 1 A block diagram showing a DDM MIMO automotive radar system.

[0053] Figure 2 A timing diagram depicting a signal sequence that can be used to determine the timing of signals transmitted by a DDM-MIMO radar system.

[0054] Figure 3 A graph depicting the Doppler spectrum of a uniform DDM-MIMO radar system as a function of the solitary frequency (and Doppler shift).

[0055] Figure 4 A flowchart depicting a method for processing raw data corresponding to sampled return signals, the sampled return signals corresponding to reflections of transmitted signals reflected by objects in the environment surrounding a DDM-MIMO radar system.

[0056] Figure 5 An example of a range-Doppler matrix of coherent integration is shown.

[0057] Figure 6 A graph depicting the Doppler spectrum of a uniform DDM-MIMO radar system as a function of the solitary frequency (and Doppler shift), where there is signal overlap of two objects within a particular Doppler bin.

[0058] Figure 7 A flowchart depicting a method for detecting Doppler spectrum overlap in a DDM MIMO radar system according to the present disclosure.

[0059] Figure 8Depicts the geometric arrangement of transmit and receive antennas in an example radar system.

[0060] Figure 9 Depicted are example spectra that may be generated for a first subject and a number of transmit channels.

[0061] Figure 10 Depicted are example spectra that may be generated for a second object for a number of transmit channels. DETAILED DESCRIPTION

[0062] The following detailed description is merely illustrative in nature and is not intended to limit the embodiments of the subject matter of the application and the use of such embodiments. As used herein, the words "exemplary" or "example" mean "serving as an example, instance, or illustration". Any embodiment or embodiment described herein as exemplary or example is not necessarily to be construed as being preferred or advantageous over other embodiments. In addition, it is not intended to be bound by any explicit or implicit theory presented in the aforementioned technical field, background technology, or the following detailed description.

[0063] Automotive radar systems are used to support advanced driver assistance systems (ADAS) features such as assisted cruise control, emergency braking, blind spot monitoring and alerts.

[0064] Various embodiments described herein relate to a Doppler division multiplexing (DDM) multiple-input multiple-output (MIMO) radar system with integrated object or obstacle tracking capabilities, including the ability to determine the direction of arrival (DOA) of detected objects.

[0065] In one or more embodiments, an example radar system usable in automotive applications includes a microcontroller unit (MCU) having a signal processor configured to generate a range-Doppler data set from raw analog-to-digital (ADC) samples such as those derived from return signals (i.e., reflections), the return signals corresponding to reflections of transmitted signals (e.g., chirps) transmitted via a transmitter (TX) antenna element of the radar system (e.g., according to a Doppler partitioning scheme having, for example, a uniform pulse repetition interval (PRI)), wherein the return signals are received via a receiver (RX) antenna element of the radar system. For example, the range-Doppler data set can be generated by performing range compression (e.g., in the form of a fast Fourier transform (FFT)) and Doppler compression (e.g., in the form of a slow-time FFT) on the raw ADC samples of one or more ADC outputs of a receiver module of the radar system. The signal processor can then perform per-channel direction-of-arrival (DOA) estimation by performing a channel-dimension FFT on each range-Doppler cell of the range-Doppler data set to produce a corresponding FFT output matrix for each range bin. The signal processor can then perform DDM decoding based on the FFT output matrix to produce a decoded range-Doppler bitmap (RDBM), where each row of the RDBM corresponds to a respective FFT output matrix of the FFT output matrix and thus corresponds to a respective range bin of the range-Doppler data set. While one or more embodiments described herein relate to the use of FFTs, it should be understood that this is intended to be illustrative and not limiting. According to one or more other embodiments, other suitable techniques for performing a discrete Fourier transform (DFT) can alternatively be used.

[0066] Standard DDM MIMO techniques mean that all radar system transmitters are active in the same time frame, where each transmitter has a respectively different self-induced chirp-to-chirp phase rotation while maintaining a common and constant pulse repetition frequency (PRF). This produces a sequence of signals that are orthogonal in the Doppler domain transmitted from each transmitter module such that, depending on the transmitter module (or TX channel) associated with the return signal, the return signals received from each object are replicated at different speeds in the Doppler domain. The DDM code determines the manner in which these replications occur in the Doppler domain, and the order of the replications needs to be known in order to correctly construct the MIMO array. Finding the order of such replicated signals in the Doppler domain can be referred to as "decoding", which is equivalent to finding the unambiguous velocity of the corresponding object within the unambiguous Doppler range of ± where T R is the pulse repetition interval (PRI).

[0067] Existing solutions for DDM decoding rely on the use of detection algorithms that are vulnerable to channel overlap errors where various Doppler velocities present in the Doppler domain separate the replicated signals. The likelihood that such values may overlap (and thus produce incorrect values) can increase, for example, in the case of objects at close range and / or higher speeds. The overlapping object signals can thus affect the driving assistance decisions of a vehicle radar system.

[0068] To mitigate these signal overlap errors, the present disclosure provides a vehicle radar system including a signal processing chain configured to detect potential overlaps occurring within a range-Doppler data frame. Specifically, the present disclosure provides a radar system signal processor configured to process a receiver channel RX linear subarray for each available transmit channel TX, determine a spectrum for each subarray, and based on those subarray spectra, identify the subarrays associated with transmitter / receiver pairs representing outliers that may have been corrupted by Doppler signal overlap. Those subarrays can then be zeroed out or otherwise modified to mitigate the effect of that corruption on the final DOA estimate of the vehicle radar system.

[0069] Figure 1 FIG. shows a block diagram of a DDM MIMO automotive radar system 100 including a DDM MIMO radar device 102 connected to a radar microcontroller unit (MCU) 104. In one or more embodiments, the device 102 can be a linear frequency modulation (LFM) DDM MIMO radar device. In one or more embodiments, the DDM MIMO radar device 102 can be embodied as a line replaceable unit (LRU) or modular component designed to be quickly replaceable in an operating position. Similarly, the radar MCU 104 can be embodied as a line replaceable unit (LRU) or modular component. Although a single or monostatic DDM MIMO radar device 102 is shown, it should be understood that additional distributed radar devices can be used to form a distributed or multistatic radar. Additionally, the depicted radar system 100 can be implemented in integrated circuit form, where the DDM MIMO radar device 102 and the radar MCU 104 are formed as separate integrated circuits (chips) or a single chip, depending on the application.

[0070] The radar device 102 includes transmit antenna elements 126 (sometimes referred to herein as "transmit antennas 126") and receive antenna elements 142 (sometimes referred to herein as "receive antennas 142") connected to a radio frequency (RF) transmitter (TX) module 118 and a receiver (RX) module 128, respectively. Each transmit antenna 126 and TX module can be designated as TX1, TX2, TX3, …… TX iassociated with a corresponding transmit channel in a set of transmit channels, where "i" is the total number of transmit (TX) channels. Each receive antenna 142 and RX module 128 can be associated with a corresponding receive channel in a set of receive channels designated herein as RX1, RX2, RX3, …… RX j associated with a corresponding receive channel in a set of receive channels, where "j" is the number of receive (RX) channels. As a non-limiting example, a radar device (e.g., radar device 102) can include individual antenna elements (e.g., antenna element 126) connected to four transmitter modules (e.g., transmitter module 118) and sixteen receiver modules (e.g., receiver module 128), respectively. The number of transmitter and receiver antenna elements and modules is intended to be illustrative rather than limiting, where in one or more other embodiments, other numbers of these elements are possible, e.g., four transmitter modules 118 and six receiver modules 128, or a single transmitter module 118 and / or a single receiver module 128. The radar device 102 includes a chirp generator 116 configured to supply a chirp input signal to the transmitter module 118. To this end, the chirp generator 116 is configured to receive input program and control signals from the MCU 104 via a digital-to-analog converter 114. As a non-limiting example, the input program and control signals include a reference local oscillator (LO) signal, a chirp start trigger signal, and a program control signal. The chirp generator 116 is configured to generate a chirp signal and send the chirp signal to the transmitter module 118 for transmission via the transmit antenna element 126. In one or more embodiments, each transmitter module includes a phase rotator 120 (sometimes referred to herein as "phase shifter 120"), which is configured to apply phase encoding to the chirp signal, where the phase rotator 120 is controlled by a program control signal generated by the MCU 104. In one or more embodiments, the phase rotator 120 can provide a uniform phase shift between the transmitter modules 118. According to one or more other embodiments, alternatively, the phase rotator 120 can provide each transmitter module 118 with the ability to implement progressive phase shifting using a non-uniform coding technique. Each transmitter module 118 includes an RF conditioning module 122 configured to filter the phase-encoded chirp signal.

[0071] Each transmitter module 118 includes a power amplifier 124 configured to amplify the filtered phase - encoded chirp signal before it is provided to and transmitted via one or more corresponding transmit antenna elements 126. By using each transmit antenna 126 to transmit a sequence of progressively phase - shifted chirp signals, each transmitter module 118 operates in Doppler division multiplexing fashion with other transmitter modules in the transmitter module 118 because these transmitter modules are programmed to simultaneously transmit the same waveform on phase - separated schedules. In this document, the transmitted chirp signal is sometimes referred to as the "transmitted signal".

[0072] In the environment of the radar device 102, the radar signal transmitted by the transmitter antenna module 118 can be reflected by an object or obstacle (which is sometimes referred to as a "target object" in the field of automotive radar systems, where the target object is a specific object for which the radar system is collecting relevant information to provide appropriate driving assistance), and a portion of the reflected radar signal, sometimes referred to in this document as the "return signal" or "reflection", is received by the receiver antenna module 128 at the radar device 102. At each receiver module 128, the received (radio - frequency) antenna signal is amplified by a low - noise amplifier (LNA) 140 and then fed to a mixer 138, where the received (radio - frequency) antenna signal is mixed with the transmitted chirp signal generated by the RF conditioning module 122. The resulting intermediate - frequency signal is fed to a high - pass filter (HPF) 136. The resulting filtered signal is fed to a variable - gain amplifier 134, which amplifies the signal before feeding it to a low - pass filter (LPF) 132. This re - filtered signal is fed to an analog - to - digital converter (ADC) 130 and output as a digital signal by each receiver module 128 (e.g., output to the signal processor 110 of the MCU 104). In this way, the receiver module 128 compresses various delayed target echoes into a plurality of sinusoidal frequency chirps, the frequencies of which correspond to the round - trip delay of the echoes.

[0073] In radar system 100, radar MCU 104 can be connected and configured to supply an input control signal to radar device 102 and receive a digital output signal generated by receiver module 128 therefrom. In one or more embodiments, radar MCU 104 includes radar controller 108 and signal processor 110, wherein either or both of the radar controller 108 and the signal processor 110 can be embodied as a microcontroller unit (MCU) or other processing unit. Radar controller 108 can receive data from radar device 102 (e.g., from receiver module 128) and can control radar parameters of radar device 102 via DAC 114, such as the frequency band, length, etc. of each radar frame. For example, DAC 114 can be used to adjust the radar chirp signal output from chirp generator 116 included in radar device 102. Signal processor 110 can be configured and arranged for signal processing tasks, such as but not limited to object identification, calculation of the range or distance to an object, calculation of the radial velocity of an object, and calculation of the DOA of the signal reflected by an object, etc. Herein, the term "DOA" or "direction of arrival" (sometimes referred to as "AoA" or "angle of arrival") refers to the angle of a signal (e.g., a radar signal) reflected by an object in the environment. Signal processor 110 can provide the calculated values associated with such calculations to storage device 112 and / or other systems via interface 106.

[0074] As a non-limiting example, interface 106 can enable MCU 104 to communicate with other systems via local area networks and wide area networks, the Internet, automotive communication buses, and / or other types of wired or wireless communication systems. In one or more embodiments, MCU 104 can provide the calculated values to other systems via interface 106, such as a radar-camera-lidar fusion system; an autonomous driving assistance system including parking, braking, or lane change assistance features; and so on. Storage device 112 can be used to store instructions for MCU 104, data received from radar device 102, calculated values from signal processor 110, etc. Storage device 112 can be any suitable storage medium, such as volatile or non-volatile memory.

[0075] To control the transmitter module 118, the radar controller 108 may be configured, for example, to generate transmitter input signals such as programs, control triggers, reference LO signals, calibration signals, spectrum shaping signals (e.g., ramp generation in the case of FMCW radar). The radar controller may be configured, for example, to receive data signals, sensor signals, and / or register programming or state machine signals for the RF (radio frequency) circuit enable sequence. In one or more embodiments, the radar controller 108 may be configured to program the transmitter module 118 with the transmitter input signals to operate in DDM mode by progressively phase-shifting the LFM chirps to be transmitted by the transmit antenna elements 126. In a selected embodiment, the radar controller 108 is configured to progressively phase-shift the LFM chirps output by the chirp generator 116 with co-prime coding (CPC) coding by programming the programmable phase rotator 120 before transmission. Each transmitter module 118 transmits a transmit channel signal of an LFM signal having a different CPC coding generated using the programmable slow-time phase rotator 120, such that the receiver module 128 may condition the object return signal (i.e., the return signal corresponding to the reflection of the transmit signal transmitted via the transmit antenna elements 126 from one or more objects, where the return signal is then received via the receive antenna elements 142) to produce a digital domain signal, which is processed by the radar MCU 104 to separate and identify the CPC-coded transmit channel signals.

[0076] At each receiver module 128, a digital output signal is generated from the target return signal for digital processing by the signal processor 110 to construct and accumulate a multiple-input multiple-output (MIMO) array vector output to form a MIMO aperture for calculating a plot or map for DOA estimation and object trajectory. Specifically, in the signal processor 110, the digital output signal may be processed by one or more fast Fourier transform (FFT) modules or discrete Fourier transform (DFT) modules, such as a fast-time (range) FFT module that produces a range chirp antenna cube (RCAC) and a slow-time (Doppler) FFT module that produces a range-Doppler antenna cube. Each channel DOA estimation module of the signal processor 110 may then perform a corresponding channel dimension FFT on each range partition of the range-Doppler antenna cube to produce a number of FFT or DFT output matrices (i.e., one FFT or DFT output matrix for each range partition of the range-Doppler antenna cube). The DDM decoder module of the signal processor 110 may then produce a decoded range-Doppler bitmap (RDBM) based on the FFT or DFT output matrices, as will be described in more detail below. The decoded RDBM may be further processed by the signal processor 110 to construct a DDM MIMO array vector, and the signal processor 110 then processes the DDM MIMO array vector to perform DOA estimation and object tracking. The MCU 104 may then output the resulting object trajectory (e.g., via the interface 106) to other automotive computing devices or user interface devices for further processing or display.

[0077] For context understanding of the operation of the DDM MIMO radar system of the radar system 100, such as Figure 1 is now referred to the Figure 2 showing a timing diagram 200, which shows linear chirp transmission schedules 201, 202, 203, 204 of four transmitters using a uniform Doppler partitioning scheme (e.g., in one or more embodiments, each transmitter corresponds to a Figure 1 transmitter module 118 and transmit antenna element 126). Each transmitter is associated with a corresponding transmit channel in the transmit channels TX1 - TX4 and is programmed to simultaneously transmit a sequence of DDM linear chirp waveforms in a single radar transmission frame. Each transmitter transmits a linear chirp waveform at a fixed and uniform pulse repetition frequency (PRF) rate (e.g., 201A, 201B), where the PRF rate is the reciprocal of the pulse repetition interval (e.g., PRF = PRI -1 ) or the reciprocal of the chirp interval time (e.g., PRF = CIT -1) Additionally, each transmitter encodes each chirp with an additional progressive phase offset by using a phase rotator in the front-end circuit. As a result of the progressive phase offset encoding, each received chirp corresponding to the reflection of the radar signal from each different transmitter effectively has a different zero radial velocity Doppler frequency shift, and individual object detections can be associated with the corresponding transmitter from which the reflected radar signal originated, which is necessary for the correct operation of subsequent MIMO virtual array construction.

[0078] Based on the following equation, the position of zero radial velocity is controlled by the progressive phase offset applied to the chirps of each transmitter:

[0079]

[0080] where f zrv,i is the zero radial velocity Doppler frequency shift of transmitter i, and where A i is the progressive phase shift (in radians) between two adjacent chirps. For example, in a 4-TX DDM MIMO radar system, each transmitter can be assigned the following progressive phase offsets and thus assigned zero radial velocity frequencies:

[0081] TX1: A1 = 0 → f zrv,1 = 0 [Hz];

[0082]

[0083] In this example, this result is depicted as follows: for transmit channel TX1, the first transmitter encodes its chirp waveform with a progressive phase offset of 0 degrees; for transmit channel TX2, the second transmitter encodes its chirp waveform with a progressive phase offset of 90 degrees; for transmit channel TX3, the third transmitter encodes its chirp waveform with a progressive phase offset of 180 degrees; and for transmit channel TX4, the fourth transmitter encodes its chirp waveform with a progressive phase offset of 270 degrees. In a DDM MIMO radar system using uniform Doppler binning, an object can be unambiguously associated with the correct transmitter and TX channel by determining the position of the object within the sectorized Doppler spectrum, provided that the object has a true Doppler offset within ±PRF / (2N ch ) before applying DDM, where N ch is the number of DDM transmitter channels.

[0084] To illustrate the principle of Doppler peak detection and transmitter association, now refer to Figure 3 , which depicts a graph of the Doppler spectrum as a function of the radian frequency (and Doppler frequency shift) of a uniform DDM-MIMO radar system (e.g., Figure 1Example embodiment of the radar system 100), where the maximum radial velocity of the object does not exceed the allocated spectral budget, thus generating an unambiguous association of the target peak and the transmitting antenna. As depicted, each transmitter channel TX1 - TX4 has a corresponding allocated spectral segment 311, 312, 313, 314 that is actually centered at the corresponding zero radial velocity frequency (e.g., 0, π / 2, π, and 3π / 2). Thus, for each object measurement 301, 302, 303, 304 with a true Doppler shift within ±PRF / (2Nch) of the zero radial velocity frequency for each transmitter channel TX1 - TX4, it can be unambiguously associated with the correct transmitter by determining the position of the object measurement within the sectorized Doppler spectrum 300. This result holds when the radial velocity of the object falls within the spectral budget segments 311, 312, 313, 314 allocated for each transmitter channel TX1 - TX4. As shown, the spectral budget segment 311 of transmitter channel TX1 can be divided into segment 311A that occurs at the start of the time period and segment 311B that occurs at the end of the time period.

[0085] In the case of an existing monolithic microwave integrated circuit (MMIC) for a fast chirp automotive radar front end, the fastest chirp signal is limited by a chirp interval time (CIT) of no less than about 15 microseconds. Thus, when a vehicle is in a dynamic driving scenario where other cars typically can travel at about 100 km / h, the range of the Doppler spectral bandwidth is sufficient for only a very few transmitters. Although more transmitters and DDM transmitter channels can be supported by significantly shortening the CIT to the microsecond level without ambiguity, it should be noted that given such high - dynamic driving conditions, such ultra - short chirps increase the cost and complexity of the radar system.

[0086] Figure 4 Illustrates an illustrative process flow of method 400 according to the present disclosure, by which the raw data corresponding to the sampled return signal corresponds to the reflection of the transmitted signal (e.g., a chirp transmitted according to a DDM scheme, e.g., Figure 2 the chirp shown in the timing diagram 200) by objects in the environment around a DDM - MIMO radar system (e.g., a vehicle radar system). As described herein, method 400 is configured to detect signal overlap occurring with the received signal of the radar system, and the received signal is processed to generate RDBM.

[0087] According to one or more embodiments, method 400 may be executed using either or both of the radar controller and the signal processor of a radar MCU. Refer to Figure 1The components of the radar system 100 describe method 400. However, it should be understood that this is illustrative and not restrictive, at least because in one or more other embodiments, other suitable radar systems may be used to perform method 400.

[0088] At block 402, the signal processor 110 of the MCU 104 of the radar system 100 receives raw sample data (sometimes referred to as "ADC samples") from one or more ADCs 130 of the receiver module 128. These ADC samples may be output as digital signals by the ADCs 130. These ADC samples represent the received signal (sometimes referred to as the "reflected signal" or "return signal") corresponding to the reflection of the transmitted signal (i.e., chirp) transmitted by the transmitter module 118 via the transmit antenna element 126 (e.g., according to the DDM scheme, such as Figure 2 the DDM scheme shown in the timing diagram 200), where the transmitted signal is reflected from one or more objects (e.g., other vehicles, trucks, road obstacles) in the environment of the radar system 100. The return signal is received by the receiver module 128 via the receive antenna element 142.

[0089] At block 404, the signal processor 110 performs range compression of the raw ADC samples, e.g., by performing a fast-time FFT or DFT on the raw ADC samples. The signal processor 110 may produce a range chirp antenna cube (RCAC) as the output of this fast-time FFT or DFT.

[0090] At block 406, the signal processor 110 performs Doppler compression of the RCAC, e.g., by performing a slow-time FFT or DFT on the RCAC. The signal processor 110 may produce a range-Doppler antenna cube as the output of this slow-time FFT or DFT. The range-Doppler antenna cube may be a three-dimensional array of dimensions m×n×p, where m represents the number of Doppler bins, n represents the number of range bins, and p represents the number of RX channels represented in the range-Doppler antenna cube (e.g., corresponding to the number of receive antenna elements 142 or the number of receiver modules 128). Herein, in the context of the range-Doppler antenna cube and the range bin matrix extracted therefrom, p is sometimes referred to as the "channel dimension".

[0091] Each element of the range-Doppler antenna cube can encode the complex amplitude of the received signal for a specific range bin, Doppler bin, and RX channel. In one or more embodiments, a given RX channel slice of the range-Doppler antenna cube can be represented as a two-dimensional (2D) array of range-Doppler cells, where the complex amplitude of each cell corresponds to the average complex amplitude within the region boundaries in Cartesian space via corresponding pairs of isodoppler lines and isorange lines. That is, the boundaries of each range bin are defined by a corresponding pair of isorange lines, and the boundaries of each Doppler bin are defined by a corresponding pair of isodoppler lines. Each range-Doppler cell corresponds to a corresponding pair of Doppler bin and range bin.

[0092] Because the transmitter module 118 simultaneously transmits chirps in the DDM-MIMO radar system, the information from all TX channels is represented in each element of the range-Doppler antenna cube. Each range bin represents the distance or separation between the radar system and an object. Each Doppler bin represents a Doppler shift value corresponding to the speed at which the object may be traveling (e.g., the ego-velocity relative to the radar system 100). Such speed can be calculated based on the determined Doppler shift associated with the object.

[0093] At block 408, the signal processor 110 performs per-channel (e.g., per RX channel) DOA estimation using the range-Doppler antenna cube. In one or more embodiments, the signal processor 110 performs per-channel DOA estimation by: extracting range bins from the range-Doppler antenna cube and performing a channel dimension FFT or DFT on each range-Doppler cell of the extracted range bins to produce an FFT or DFT output matrix (i.e., where each FFT or DFT output matrix corresponds to a corresponding range bin of the range-Doppler antenna cube).

[0094] At block 410, the signal processor 110 performs DDM decoding of the range-Doppler antenna cube based on the FFT or DFT output matrix generated by the per-channel DOA estimation performed at block 408 to produce a decoded RDBM. At block 411, the RDBM is processed as described herein to determine whether the RDBM encodes values associated with overlapping signals.

[0095] At block 412, the signal processor 110 constructs a virtual MIMO array based on the decoded RDBM and the range-Doppler antenna cube. At block 414, the signal processor 110 uses the virtual MIMO array to generate object location data based on the DOA estimation. At block 416, the signal processor 110 generates object tracking data based on the object location data.

[0096] During operation, the signal processor 110 of the radar system 100 generates a range bin matrix from a range-Doppler antenna cube (e.g., generated at block 404 above), and performs a channel dimension FFT on each range-Doppler cell of the extracted range bin matrix to generate an FFT output matrix. The FFT output matrix is used to generate a Doppler bin maximum array that represents the maximum value of each column detected in the FFT output matrix. The maximum arrays can then be combined to generate a coherent integrated range-Doppler matrix (RDM). Herein, the RDM may differ from the RDBM in that the RDBM includes a matrix or 2D array of binary values (e.g., 0 and 1), while the RDM may include a matrix or 2D array of signal amplitudes. Figure 5 An example of such a coherent integrated RDM 500 is shown. In the coherent integrated RDM 500, data from each TX channel is encoded according to the corresponding DDM code for each TX channel such that each TX channel data set is represented in the RDM 500 and offset by a predetermined number of Doppler bins, where the offset is different for each TX channel (e.g., as defined by the corresponding DDM code). In Figure 5 the multiple Doppler bins 502 depicted in, data corresponding to different transmitters are represented. For example, data representing a velocity of approximately -30 m / s corresponding to the first TX channel TX1 is set in the first Doppler bin 502. Data representing a velocity of approximately -28 m / s corresponding to the second TX channel TX2 is set in the second Doppler bin 502.

[0097] In an ideal coherent integrated RDM, there is sufficient Doppler offset between the various multiple Doppler bins such that the signals associated with the different transmit channels depicted therein do not overlap with each other. However, in situations involving higher velocity objects (e.g., other vehicles), signals that typically do not overlap with signals in other TX channels and are constrained to a specific Doppler bin by the RDM associated with a particular TX channel may spread and overlap, thereby corrupting signals associated with other TX channels in other Doppler bins.

[0098] This phenomenon is depicted in Figure 6 which depicts an RDM 600 of coherent integration of received radar signals for processing by a 6 transmitter, 8 receiver radar system for an environment with two objects. As shown, in the Doppler dimension (i.e., along the horizontal axis), the signals 602, 604, 606, 608, 610, and 612 associated with the six transmit channels for the first object ("Object 1") are all separated from each other. Similarly, in the Doppler dimension (i.e., along the horizontal axis), the signals 622, 624, 626, 628, 630, and 632 associated with the six transmit channels for the second object ("Object 2") are all separated from each other

[0099] However, as also shown in Figure 6 , the signals associated with the first transmission channel TX1 and the first object (i.e., signal 602) overlap in the Doppler domain with the signals associated with the second transmission channel TX2 and the second object (i.e., signal 622). Based on Figure 6 the RDM data depicted in

[0100] , this overlap can interfere with the accurate determination of the DOA of object 1 and / or object 2. Figure 6 To further illustrate

[0101] Tx1 Tx2 Tx3 Tx4 Tx5 Tx6 Object 1 6 950 838 662 454 214 Object 2 86 6 918 742 534 294

[0102] Table 1

[0103] To illustrate this signal overlap mathematically, the ideal steering vectors for objects 1 and 2 can be expressed as follows, where v1 is the ideal steering vector of the received signal associated with object 1 and v2 is the ideal steering vector of the received signal associated with object 2.

[0104]

[0105] In equations (1) and (2), the element represents the complex signal value for transmission channel m, reception channel n, and object j (e.g., from the RDM data generated at Figure 4 box 410). Thus, in this example (i.e., involving two objects), if the signal associated with transmission channel TX1 and object 1 overlaps with the signal associated with transmission channel TX2 and object 2, the updated steering vectors v1' and v2' for objects 1 and 2 with this overlap become:

[0106]

[0107]

[0108] The resulting overlap can produce significant distortion of the steering vectors associated with the two objects, as depicted by the bold modified terms in expressions (3) and (4).

[0109] This distortion can render the signals associated with the transmit channel TX1 for object 1 and the transmit channel TX2 for object 2 unavailable for reliable MIMO array construction and ultimately unavailable for DOA estimation. Detecting this overlap can be difficult, and this overlap can occur in several ways (i.e., ranging from a complete overlap where the overlapping signals are within the same Doppler bin to a partial overlap resulting from signals in adjacent Doppler bins). Detecting this overlap can become even more difficult when the overlap is caused by signals associated with three or more objects.

[0110] To mitigate this problem, the present disclosure provides a method to detect this overlap attributable to direct overlap or adjacent signals such that the overlapping data can be ignored or modified or assigned a reduced weight to account for signal overlap.

[0111] Specifically, Figure 7 is a flowchart depicting method 700 for detecting Doppler spectrum overlap in a DDM MIMO radar system according to the present disclosure. Method 700 can be implemented to perform all or a portion of block 411 described above with respect to Figure 4 According to one or more embodiments, or using other hardware components configured to process signals, method 700 can be performed using either or both of the radar controller and signal processor of the radar MCU (e.g., Figure 1 radar controller 108 and / or signal processor 110) of Figure 1 Method 700 is described with reference to the elements of radar system 100 of

[0112] Method 700 is described with respect to a specific embodiment of a radar system (e.g., radar system 100) including six transmit antennas and eight receive antennas and signal processing with respect to two objects. In this embodiment, the antennas are arranged in a specific geometric arrangement with uniform spacing that enables DDM MIMO processing of the received signals. The geometric spacing is depicted in Figure 8 Specifically, Figure 8Depict the geometric arrangement of the transmit antennas and receive antennas in an exemplary radar system. Specifically, six transmit antennas 802, 804, 806, 808, 810, and 812 are placed in a certain arrangement, where transmit antennas 802, 804, 806 are arranged in a first linear arrangement, and antennas 802, 804, 806 are generally spaced apart by a distance of two wavelengths (2*λ). Similarly, transmit antennas 808, 810, 812 are arranged in a second linear arrangement, where antennas 808, 810, 812 are generally spaced apart by a distance of two wavelengths (2*λ). As shown, the two linear arrangements of antennas 802, 804, 806 and 808, 810, 812 are spaced apart by a distance of four wavelengths (4*λ) from each other.

[0113] The receive antenna 820 is arranged in a single linear arrangement having a total length of three and a half wavelengths (3.5*λ), where the spacing between individual receive antennas 820 is half a wavelength (0.5*λ). Thus, the spacing between the various receive antennas 820 is considered uniform because each receive antenna 820 is spaced from its closest adjacent receive antenna by a distance that is uniform for all receive antennas 820. In Figure 8 the description, the wavelength distance is considered to be the wavelength of the signals configured to be transmitted and received by the antenna arrangement depicted in Figure 8 .

[0114] Returning to method 700, at block 702, receive an RDM (e.g., as the output of block 410 of method 400 described above). The RDM is a multi-dimensional array of data that includes a plurality of complex signal values derived from the reflected signals received by the receive antennas of the radar system. Each complex value in the RDM is associated with a specific Doppler bin (where the Doppler bins in the RDM are associated with specific TX channels), a range bin, and an RX channel.

[0115] At block 704, initialize an object counter to the value 1, thereby designating the first object in a set of identified objects.

[0116] At block 706, using the RDM received at block 702, use the RDM to determine a receive subarray for the current object O N . The receive linear subarray is a multi-dimensional array extracted from the RDM and includes signal data associated with a specific object on different receiver channels RX. Specifically, after signal processing (e.g., range and Doppler processing) and constant false alarm rate (CFAR) detection, the range and Doppler bins of each identified object are known. Thus, for each object, the steering vector of the object is extracted from the RDM using the object's own range and Doppler bins to generate a receive subarray.

[0117] In this particular example, the receive subarray is considered linear because the geometric spacing between the receive antennas of the exemplary radar system is uniform - all antenna elements are spaced the same distance apart from each other, as shown by the receive antennas 820 of Figure 8 In embodiments of the inventive system that incorporate a method 700 for a radar system with a non-linear receive antenna configuration, a variety of different methods can be utilized to generate a linear receive subarray for the current frame. For example, in a radar system with multiple receive antennas where a subset of the antennas are spaced evenly apart from each other, a uniform receive subarray can be constructed by extracting only the values of those evenly spaced antennas from the RBM. Alternatively, zero-padding or other advanced interpolation or reconstruction techniques can be used to generate regions of zero values in the receive subarray where receive antennas are missing or otherwise not evenly spaced from the values of other receive antennas. In method 700, it is generally preferred that the receive subarray being processed is linear because this configuration can be beneficial as it enables the performance of a one-dimensional FFT (a simpler mathematical operation than performing a multi-dimensional FFT) on the data in the linear receiver array to generate specific output data such as an azimuth spectrum or an elevation spectrum.

[0118] At block 708, after generating the uniform receive subarray, the transmit channel counter TX N is initialized to the value 1. The method then at block 710 calculates a spectrum for the target object O N and the transmit channel TX N using the uniform receive subarray. Typically, this will involve performing an FFT on the data in the uniform receive subarray associated with the transmit channel TX N In a specific embodiment, this involves determining the frequency domain representation of the data in the uniform receive subarray associated with a particular transmit channel TX, for example, by performing a fast Fourier transform (FFT) on the data. However, as is known in the art, other methods for calculating a spectrum based on the input data can be utilized, such as Welch's method, autoregressive model estimation, MUSIC (multiple signal classification), IAA (iterative adaptive approach), etc.

[0119] At block 712, a determination is made as to whether an additional transmit channel is available. If so, then at block 714, the transmit channel counter TX N is incremented and the method returns to block 710 to determine another spectrum for that transmit channel.

[0120] At the end of the control loop implemented at blocks 710, 712, and 714, a plurality of spectra have been calculated, where the spectra include, for each transmit channel TX, for the current object O NThe frequency domain representation of the data in the uniform receiving subarray. Thus, for a radar system with M transmit antennas and N transmit channels, the following M FFTs can be performed for each object according to blocks 710, 712, and 714.

[0121]

[0122] …

[0123]

[0124] For illustration, Figure 9 depicts an example spectrum that can be generated for a first object based on an RDM including signals associated with several transmit channels. In the Figure 9 example depicted, there are six graphs 902, 904, 906, 908, 910, and 912 depicting the spectra associated with six different transmit channels TX. As shown, the horizontal axis of each graph 902, 904, 906, 908, 910, and 912 depicts frequency or FFT bins, and the vertical axis of each graph 902, 904, 906, 908, 910, and 912 represents the power (in dB) of the signal associated with the corresponding frequency bin. In each spectrum, the frequency bin associated with the largest peak in the spectrum is associated with the object, while peaks with smaller power are associated with sidelobes, noise, or other signal interferences.

[0125] In a vehicle radar system application where there is no Doppler overlap between signals associated with different objects, the spectra generated by blocks 710, 712, and 714 of method 700 will typically only exhibit a single large peak indicating the frequency bin associated with a single object. Additionally, the spectra associated with each transmit channel TX will be very similar and exhibit little or no interference.

[0126] However, in this example, in the case where the processed radar signals involve two objects that generate reflected signals and the reflected signals exhibit a certain degree of Doppler overlap, it can be seen that in the Figure 9 spectrum, graph 902 depicts a spectrum different from the spectra of the remaining graphs. Specifically, graph 902 exhibits two different large peaks 950 and 952, while the other spectra only depict a single large peak. The double peaks depicted in graph 902 and not present in the other graphs 904, 906, 908, 910, and 912 indicate that there is Doppler overlap in the signal processed to generate graph 902, and the signal associated with transmit channel TX that generated this spectrum is distorted due to this Doppler overlap.

[0127] Returning to method 700, for the current object O NAfter generating the spectrum for each transmission channel, at block 712, it is determined that no additional transmission channels are reserved. The method then moves to block 716, where the spectrum generated for the current object O N is analyzed to identify outlier spectra, as described above, where the outlier spectra indicate that the signals associated with the transmission channels of the spectrum are distorted due to Doppler overlap.

[0128] The so-called identification of outlier spectra can be performed in any suitable manner to identify spectra that are very different from other spectra. In an embodiment, for example, the identification of outlier spectra can involve analyzing each spectrum to determine the number of local peaks contained in each spectrum (e.g., using a conventional peak detection algorithm, such as neighbor value comparison, where a local peak or local maximum is the position of a value that is not less than the values of two adjacent values), and then determining the spectrum that includes a number of peaks different from other spectra, which designates the outlier spectrum. Other methods for detecting outlier spectra can involve calculating the cross-correlation between spectrum pairs to identify the spectrum with the lowest cross-correlation with other spectra, which will indicate that the spectrum is an outlier. Similarly, machine learning methods can be utilized to analyze spectra to determine which spectrum in a set of spectra is an outlier based on learned model constraints.

[0129] In a specific method, each spectrum is analyzed to determine the number of local peaks present in each spectrum. For example, Table 2 below lists Figure 9 the local peak information for each of the six spectra depicted therein. Specifically, local peaks are defined where the peak has a peak magnitude higher than a specific threshold, in this example, the specific threshold is 10 decibels (dB) lower than the maximum value of the spectrum, which is shown by Figure 9 the horizontal dashed lines on graphs 902, 904, 906, 908, 910, and 912. In various embodiments, the threshold can be a predefined value selected based on the attributes of the radar system or user preferences.

[0130] Peak information Tx1 Tx2 Tx3 Tx4 Tx5 Tx6 Number of peaks 2 1 1 1 1 1 Frequency partition 43、55 57 57 57 57 57

[0131] Table 2

[0132] As shown in Table 2, there are three frequency partitions with local peaks across all spectra - partitions 43, 55, and 57. Frequency partition 43 contains a local peak only once across all spectra (i.e., in graph 902 associated with transmit channel Tx1), frequency partition 55 contains local peaks only in the spectra associated with transmit channel Tx1 (again, as depicted in graph 902), however, frequency partition 57 contains local peaks five different times in the spectra associated with transmit channels Tx2, Tx3, Tx4, Tx5, and Tx6 (i.e., as shown in graphs 904, 906, 908, 910, and 912). A threshold can then be determined, where the threshold specifies a particular number of occurrences of local peaks required on the spectrum (e.g., greater than 3 occurrences in this example) to constitute a 'true' peak of a particular transmit channel rather than a peak attributable to Doppler overlap. The occurrence threshold is a predefined value that can be selected by the user or designer of the radar system. However, in some embodiments, the threshold can be set equal to approximately 50% of the number of transmit channels in the radar system. In this case, if the number of local peaks occurring in a particular frequency partition is less than 50% of the spectra of the available transmit channels, then those local peaks can be attributes of Doppler overlap as described herein and discarded or ignored. However, if the number of local peaks occurring in a particular frequency partition is greater than 50% of the spectra of the available transmit channels, then that local peak is likely associated with the object signal rather than being attributable to Doppler overlap. In this case, those local peaks are retained and used for MIMO array processing. Thus, with a selected threshold, a true object peak can be defined as a local peak that occurs in the same frequency partition a threshold number of times in different spectra. If a particular frequency partition meets those requirements for local peaks on the spectrum, then it can be determined that the various spectra containing local peaks at that partition location are associated with non-Doppler overlap signals, that is, those spectra contain 'true' peaks. Other spectra (e.g., containing multiple local peaks found in frequency partitions not associated with local peaks in other spectra) can be designated as outliers because the peaks of those spectra are likely caused by Doppler overlap signals.

[0133] At block 718, the RDM is updated based on the identified outlier spectra. Specifically, the object O associated with the outlier spectra in the RDM NValues associated with transmit channel TX can be set to zero, or otherwise modified, to produce a corrected RDM that reduces the effects of Doppler-overlapped signals associated with those particular transmit channels TX. In other embodiments, instead of setting those values to zero, when processing the RDM to produce a corrected RDM that is used (e.g., at block 412) to produce MIMO array data for DOA estimation, producing the corrected RDM can involve assigning specific weighting values to the values in the RDM associated with that transmit channel TX to reduce the impact of those values on the final MIMO array determination, thereby reducing the impact of data values associated with Doppler-overlapped signals.

[0134] At block 720, a determination is made as to whether there are additional objects (e.g., based on the output data of block 410 of method 400). If so, then at block 722, counter O N is incremented and the method returns to block 710 to generate a new set of spectra for the new object.

[0135] In the current example where the RDM encodes data for two objects, Figure 10 example spectra that can be generated for a second object for several transmit channels are depicted. In Figure 10 the example depicted, there are six graphs 1002, 1004, 1006, 1008, 1010, and 1012 depicting spectra associated with six different transmit channels TX.

[0136] As shown, the horizontal axis of each of graphs 1002, 1004, 1006, 1008, 1010, and 1012 depicts frequency or FFT bins, and the vertical axis of each of graphs 1002, 1004, 1006, 1008, 1010, and 1012 represents the power (in dB) of the signal associated with the corresponding frequency bin. In each spectrum, the frequency bin associated with the largest peak in the spectrum is associated with the object, while peaks with smaller power are associated with sidelobes, noise, or other signal interference.

[0137] Due to signal Doppler overlap, it can be seen that in Figure 10 the spectrum of, the graph 1004 associated with transmit channel Tx2 depicts a spectrum different from the spectra depicted in the remaining graphs. Specifically, graph 1004 presents two different large peaks 1050 and 1052, while the other spectra depict only a single large peak. The presence of the double peaks depicted in graph 1004 and not present in the other graphs 1002, 1006, 1008, 1010, and 1012 indicates that there is Doppler overlap in the signal being processed to produce graph 1004, and the signal associated with transmit channel Tx2 that produced this graph is distorted due to this Doppler overlap.

[0138] Returning to method 700, thus, after generating the spectrum for all transmit channels for the current object O N (here is the second object), at block 712, no additional transmit channels are reserved and the method moves to block 716, where the spectrum generated for the second object is analyzed to identify outlier spectra.

[0139] In one embodiment, each spectrum is analyzed to determine the number of local peaks present within each spectrum. For example, Table 3 below lists Figure 10 the local peak information for each of the six spectra depicted in. Specifically, local peaks are defined where the peak has a peak magnitude higher than a specific threshold, which in this example is 10 dB lower than the maximum value of the spectrum, as shown by Figure 10 the horizontal dashed lines on graphs 1002, 1004, 1006, 1008, 1010, and 1012. In various embodiments, the threshold can be a predefined value selected based on the properties of the radar system or user preferences.

[0140] Peak information Tx1 Tx2 Tx3 Tx4 Tx5 Tx6 Number of peaks 1 2 1 1 1 1 Position 42 43、55 42 42 42 42

[0141] Table 3

[0142] As shown in Table 3, there are three frequency partitions with local peaks across all spectra - partitions 42, 43, and 55. Frequency partition 43 contains a local peak only once across all spectra (i.e., in Chart 1004 associated with transmit channel Tx2), frequency partition 45 contains a local peak only in the spectrum associated with transmit channel Tx2 (again, as depicted in Chart 1004), however, frequency partition 42 contains local peaks five different times in the spectra associated with transmit channels Tx1, Tx3, Tx4, Tx5, and Tx6 (i.e., as shown in Charts 1002, 1006, 1008, 1010, and 1012). A threshold can then be estimated for a specific number of local peak occurrences (e.g., occurrences greater than 3 times in this example). The occurrence threshold is a predefined value that can be selected by the user or designer of the radar system. However, in some embodiments, the threshold can be set equal to approximately 50% of the number of transmit channels in the radar system. In such a case, if the local peaks occurring in a specific frequency partition are less than 50% of the spectra of the available transmit channels, then those local peaks can be attributes of Doppler overlap as described herein and be discarded or ignored. However, if the local peaks occurring in a specific frequency partition are greater than 50% of the spectra of the available transmit channels, then that local peak is likely associated with an object signal rather than attributable to Doppler overlap. In such a case, those local peaks are retained and used for MIMO array processing. Thus, with a selected threshold, a true object peak can be defined as a local peak that occurs in the same frequency partition a threshold number of times in different spectra. If a frequency partition meets those requirements, then it can be determined that the spectra containing local peaks at that partition location are associated with non-Doppler overlap signals. Other spectra (e.g., containing multiple local peaks found in frequency partitions not associated with local peaks in other spectra) can be designated as outliers.

[0143] At block 718, the RDM is updated again based on the identified outliers. Specifically, the values in the RDM associated with object O N (the second object in this example) and transmit channel TX can be set to zero to produce a corrected RDM, or otherwise modified to reduce the effect of the Doppler overlap signal associated with that transmit channel TX. In other embodiments, instead of setting those values to zero, when processing the RDM to produce a corrected RDM, to (e.g., at block 412) produce MIMO array data for DOA estimation, specific weighting values can be associated with the values in the RBBM associated with that transmit channel TX to reduce the impact of those values on the final MIMO array determination, thereby reducing the impact of the data values associated with the Doppler overlap signal.

[0144] If it is determined at block 720 that there are no additional objects, the method proceeds to block 716 and ends.

[0145] Thus, at the end of method 700, for each identified object, signals associated with transmit channels that include Doppler overlap have been identified (by identifying the outlier spectra of those transmit channels TX), and the detrimental effects of that overlap can be corrected (e.g., by modifying the RDM to zero or otherwise modify the values associated with those Doppler-overlapping signals). Thus, when processing the RDM to perform DOA estimation and determine other attributes of the identified object (e.g., velocity), the determined values will be more accurate than the same estimation performed on data-coded signals that include Doppler overlap.

[0146] In some aspects, the techniques described herein relate to a radar system that includes: a plurality of transmitter modules configured to transmit a plurality of transmitted radar signals according to a Doppler domain multiplexing (DDM) scheme, where each transmitted radar signal is associated with a transmit channel among a plurality of transmit channels; a plurality of receiver modules configured to receive reflections of the plurality of transmitted signals reflected by at least one object and generate signals based on the received reflections; and a signal processor configured to: use the signals received by the plurality of receiver modules to determine a range-Doppler map that includes values associated with a plurality of range bins and a plurality of Doppler bins; determine a first sub-array that includes values associated with a first range bin of the range-Doppler map, where the first range bin is associated with a first object; use the first sub-array to calculate a plurality of spectra, where each spectrum of the plurality of spectra is associated with a transmit channel among the plurality of transmit channels; calculate an attribute of each spectrum of the plurality of spectra; use the attribute of each spectrum to determine a first spectrum among the plurality of spectra that includes a local peak not in other spectra among the plurality of spectra; and modify the values in the range-Doppler map associated with the transmit channel associated with the first spectrum to produce a corrected range-Doppler map; and use the corrected range-Doppler map to determine an estimated direction of arrival of the first object.

[0147] In some aspects, the techniques described herein relate to a radar system, where, to calculate an attribute of each spectrum of the plurality of spectra, the signal processor is configured to determine the number of local peaks within a threshold of the maximum value of each spectrum.

[0148] In some aspects, the techniques described herein relate to a radar system, where, to use the attribute of each spectrum to determine a first spectrum among the plurality of spectra that includes a local peak not in other spectra among the plurality of spectra, the signal processor is configured to determine that the number of peaks in the first spectrum is different from the number of peaks in each other spectrum among the plurality of spectra.

[0149] In some aspects, the techniques described herein relate to a radar system in which a plurality of receiver modules are arranged in a linear receiving subarray, and the first subarray is a uniform receiving subarray.

[0150] In some aspects, the techniques described herein relate to a radar system in which, in order to modify values in a range-Doppler map associated with a transmit channel associated with a first spectrum to produce a corrected range-Doppler map, a signal processor is configured to set the values in the range-Doppler map associated with the transmit channel associated with the first spectrum to zero values.

[0151] In some aspects, the techniques described herein relate to a radar system in which, in order to calculate a plurality of spectra, a signal processor is configured to calculate a fast Fourier transform of values in a first subarray.

[0152] In some aspects, the techniques described herein relate to a radar system in which, in order to calculate attributes of each of a plurality of spectra, a signal processor is configured to calculate cross-correlation values of spectrum pairs in the plurality of spectra.

[0153] Although examples have been described with reference to automotive radar systems, the systems and methods described herein may be implemented in conjunction with other types of radar systems.

[0154] The foregoing detailed description is merely illustrative in nature and is not intended to limit the embodiments of the subject matter or the application and use of such embodiments.

[0155] As used herein, the word "exemplary" means "serving as an example, instance, or illustration." Any embodiment described herein as exemplary is not necessarily to be construed as preferred or advantageous over other embodiments. Further, there is no intention to be bound by any theory, whether explicit or implicit, presented in the foregoing technical field, background, or detailed description.

[0156] The connecting lines shown in the various figures contained herein are intended to represent example functional relationships and / or physical couplings between various elements. It should be noted that there may be many alternative or additional functional relationships or physical connections in embodiments of the subject matter. Additionally, certain terms may be used herein only for reference purposes and thus are not intended to be limiting, and unless the context clearly indicates otherwise, numerical terms such as "first," "second," and other such terms referring to structures do not imply an order or sequence.

[0157] As used herein, "node" means any internal or external reference point, connection point, junction point, signal line, conductive element, etc., at which a given signal, logic level, voltage, data pattern, current, or quantity exists. Additionally, two or more nodes may be implemented by a single physical element (and two or more signals may be multiplexed, modulated, or otherwise distinguished even if received or output at a common node).

[0158] The foregoing description refers to elements or nodes or features being "connected" or "coupled" together. As used herein, unless stated otherwise explicitly, "connected" means that an element is directly connected to another element (or in direct communication with another element), and not necessarily in a mechanical manner. Similarly, unless stated otherwise explicitly, "coupled" means that an element is directly or indirectly connected to another element (or in direct or indirect communication with another element, electrically or otherwise), and not necessarily in a mechanical manner. Thus, while the schematic diagrams shown in the figures depict an exemplary arrangement of elements, additional intervening elements, devices, features, or components may be present in embodiments of the subject matter being depicted.

[0159] Although at least one exemplary embodiment has been presented in the foregoing detailed description, it should be understood that there are a vast number of variations. It should also be understood that the exemplary embodiments or exemplary embodiments described herein are not intended to limit in any way the scope, applicability, or configuration of the claimed subject matter. Indeed, the foregoing detailed description will provide those skilled in the art with a convenient guide for implementing the described embodiments or exemplary embodiments. It should be understood that various changes may be made to the functionality and arrangement of the elements without departing from the scope defined by the claims, which scope includes known equivalents and foreseeable equivalents at the time of filing this patent application.

Claims

1. A radar system, characterized in that: include: a plurality of transmitter modules configured to transmit a plurality of transmitted radar signals according to a Doppler domain multiplexing (DDM) scheme, wherein each transmitted radar signal is associated with a transmission channel of a plurality of transmission channels; a plurality of receiver modules configured to receive reflections of the plurality of transmitted signals reflected by at least one object and to generate signals based on the received reflections; as well as A signal processor configured to: determining a range-Doppler map using signals received by the plurality of receiver modules, the range-Doppler map comprising values ​​associated with a plurality of range bins and a plurality of Doppler bins, determining a first subarray comprising values ​​associated with a first range partition of the range-Doppler map, wherein the first range partition is associated with a first object; calculating a plurality of frequency spectra using the first subarray, wherein each frequency spectrum in the plurality of frequency spectra is associated with a transmit channel in the plurality of transmit channels; calculating a property of each spectrum in the plurality of spectrums; determining a first spectrum of the plurality of spectra that includes a local peak not present in other spectra of the plurality of spectra using the properties of each spectrum; modifying values ​​in the range-Doppler map associated with the transmit channels associated with the first spectrum to determine a corrected range-Doppler map; and An estimated direction of arrival of the first object is determined using the corrected range-Doppler map.

2. The radar system according to claim 1, characterized in that To calculate the property of each spectrum of the plurality of spectra, the signal processor is configured to determine a number of local peaks in each spectrum that are within a threshold of a maximum value of the spectrum.

3. The radar system according to claim 2, characterized in that In order to determine a first spectrum of the multiple spectra that includes local peaks that are not in other spectra of the multiple spectra using the properties of each spectrum, the signal processor is configured to determine that the number of peaks in the first spectrum is different from the number of peaks in each other spectrum of the multiple spectra.

4. The radar system according to claim 1, characterized in that The plurality of receiver modules are arranged into a linear receiving subarray, and the first subarray is a uniform receiving subarray.

5. The radar system according to claim 1, characterized in that To modify the values ​​in the range-Doppler map associated with the transmission channel associated with the first spectrum to determine a corrected range-Doppler map, the signal processor is configured to set the values ​​in the range-Doppler map associated with the transmission channel associated with the first spectrum to zero values.

6. The radar system according to claim 1, characterized in that To calculate the plurality of spectra, the signal processor is configured to calculate a Fast Fourier Transform of the values ​​in the first sub-array.

7. The radar system according to claim 1, characterized in that To calculate the property of each spectrum of the plurality of spectra, the signal processor is configured to calculate cross-correlation values ​​of pairs of spectra of the plurality of spectra.

8. A radar system, characterized in that: include: a transmitter system configured to transmit transmitted signals, wherein each transmitted signal is associated with a transmission channel of a plurality of transmission channels; a receiver system configured to receive the reflected signal; and A signal processor configured to: determining a range-Doppler map using the reflected signal, determining a first subarray comprising values ​​associated with a first range partition of the range-Doppler map; calculating a plurality of frequency spectra using the first subarray, wherein each frequency spectrum in the plurality of frequency spectra is associated with a transmit channel in a plurality of transmit channels; determining that the first spectrum among the plurality of spectra includes a local peak that is not in other spectra among the plurality of spectra by detecting a local peak in the first spectrum among the plurality of spectra; and Values ​​in the range-Doppler map associated with the transmit channels associated with the first spectrum are modified to determine a corrected range-Doppler map.

9. The radar system according to claim 8, characterized in that To determine that the first spectrum includes local peaks not present in other spectra of the plurality of spectra, the signal processor is configured to determine that a number of peaks in the first spectrum is different from a number of peaks in each other spectrum of the plurality of spectra.

10. A method, characterized in that include: determining a range-Doppler map using signals received by the plurality of receiver modules, the range-Doppler map comprising values ​​associated with a plurality of range bins and a plurality of Doppler bins; determining a first subarray comprising values ​​associated with a first range partition of the range-Doppler map, wherein the first range partition is associated with a first object; calculating a plurality of frequency spectra using the first subarray, wherein each frequency spectrum in the plurality of frequency spectra is associated with a transmit channel in the plurality of transmit channels; calculating a property of each spectrum in the plurality of spectrums; determining a first spectrum of the plurality of spectra that includes a local peak not present in other spectra of the plurality of spectra using the properties of each spectrum; modifying values ​​in the range-Doppler map associated with the transmit channels associated with the first spectrum to determine a corrected range-Doppler map; and An estimated direction of arrival of the first object is determined using the corrected range-Doppler map.

Images